Risk, trust, and altruism in genetic data sharing

• Published date: 01 December 2023
• Author: Zeeshan Samad,Myrna Wooders,Bradley Malin,Yevgeniy Vorobeychik
• Date: 01 December 2023
• DOI: http://doi.org/10.1111/jpet.12678

Received: 22 August 2023

|

Accepted: 6 November 2023

DOI: 10.1111/jpet.12678

ORIGINAL ARTICLE

Risk, trust, and altruism in genetic data

sharing

Zeeshan Samad

1

|Myrna Wooders

2

|Bradley Malin

3

|

Yevgeniy Vorobeychik

4

1

Department of Economics and Finance,

Utah State University, Logan, Utah, USA

2

Department of Economics, Vanderbilt

University, Nashville, Tennessee, USA

3

Department of Biomedical Informatics,

Vanderbilt University Medical Center,

Nashville, Tennessee, USA

4

Department of Computer Science and

Engineering, Washington University in

St Louis, St Louis, Missouri, USA

Correspondence

Zeeshan Samad, Department of

Economics and Finance, Utah State

University, Logan, UT, USA.

Email: zeeshan.samad@usu.edu

Funding information

National Institutes of Health

Abstract

How does concern about genetic data privacy compare

with other concerns? We conduct behavioral experi-

ments to compare risk attitudes towards sharing

genetic data with a healthcare provider with risk

attitudes towards sharing financial data with a money

manager. Both scenarios involve identical decisions

and monetary stakes, permitting us to focus on how the

framing of data sharing influences attitudes. To delve

deeper into individual motivations to share data, we

provide treatments that study how data sharers'

altruism and trust affect their decisions. Our findings

(with 162 subjects) indicate that individuals are more

willing to risk a loss to privacy of genetic data (for an

anticipated return framed as health benefits) than they

are to risk loss of financial data (for an anticipated

return in financial benefits). We also find that 50%–60%

of data recipients choose to protect another person's

data, with no significant differences between frames.

KEYWORDS

data privacy, experiment, genetics

1|INTRODUCTION

Cheap DNA sequencing has enabled broad collection, processing, and sharing of genetic

information (Van Dijk et al., 2018). This information is disclosed by patients to physicians, by

research participants to scientists, and by general consumers to direct‐to‐genetic testing

companies (e.g., 23andme.com and ancestry.com) for various purposes. Genetic information

J Public Econ Theory. 2023;25:1251–1269. wileyonlinelibrary.com/journal/jpet © 2023 Wiley Periodicals LLC.

|

1251

sharing provides an individual with the opportunity to improve health and contribute to

societal endeavors. At the same time, while some people share genetic information openly (e.g.,

posting to websites, such as OpenSNP or GEDMatch), many others consider such information

to be highly personal and potentially sensitive (Greshake et al., 2014; Greytak et al., 2018;

Haeusermann et al., 2017). As a result, many people prefer that their information be managed

and used in a manner that preserves their expectation of privacy (Erlich & Narayanan, 2014).

However, sharing data with another party is not without risks. For instance, a data recipient

may fail to protect a data sharer's privacy through weak anonymization practices or misuse the

data due to lack of access control and institutional oversight.

Most investigations that examine the extent to which people are concerned about privacy

and security of their genetic data elicit people's privacy concerns through surveys (Botkin

et al., 2012; Clayton et al., 2018; Condit et al., 2016; Duquette et al., 2012; Edwards et al., 2012;

Kaufman et al., 2009; Lemke et al., 2010; Rauscher et al., 2015; Sanderson et al., 2017). Given

that survey instruments sometimes fall prey to the privacy paradox, whereby reported attitudes

about privacy end up being inconsistent with measures people actually take to protect their

data (Norberg et al., 2007), one would be justified to question the internal validity of the studies

mentioned above.

A natural way to circumvent the privacy paradox is to elicit people's privacy concerns

through an incentive‐compatible method rather than unincentivized survey responses. To this

end, we adopt an economic perspective, wherein data sharing is modeled as a risky decision

that depends on an individual's beliefs about the potential risks and benefits associated with

data sharing. In this representation, sharing genetic data may result in a long‐term benefit, for

example, treatment for an undiagnosed condition (Bastarache et al., 2018), but may expose an

individual to privacy risks.

In this work, we investigate, by conducting laboratory experiments, the extent to which

subjects value the privacy of their genetic data relative to the privacy of their financial data. To

capture their willingness to share data, we ask them to make an incentivized risky decision. To

capture the extent to which decisions about sharing genetic data differ from decisions about

sharing financial data, we frame the risky decision in two different ways. In one frame, which

we call the genetic frame, a patient can undergo a genetic test, and thereby share genetic data

with a physician. In a second frame, which we call the investment frame, an investor can invest

in a risky asset, thereby sharing financial data with a money manager. Crucially, actual

monetary gain and loss potentials are identical in both frames. This approach is most closely

related to the experiment developed by Binder and Nuscheler (2017), who implement vaccine,

surgery, and neutral frames to assess subjects' willingness to take a risk.

In each frame, subjects, acting as patients or investors, are provided with a fixed amount of

money as an endowment. Each subject makes a decision on whether to share data, that is,

whether or not to undergo a genetic test (in the genetic frame) or whether or not to make an

investment (in the investment frame). If a subject does not share data, she keeps her

endowment as her final payout. If she does share data, she faces two uncertain possibilities, one

positive and the other negative. The positive outcome is a large reward with a small probability,

meant to represent medical treatment in the genetic frame or financial gain in the investment

frame. The negative outcome is that the subject loses her endowment with a larger probability,

representing the risk associated with a potential privacy compromise.

The design of the aforementioned experimental setting, henceforth referred to as the base

setting, effectively removes the role of the data recipient from the equation, thus casting the

problem purely as a risky decision. We expand on this experimental design through three more

1252

|

SAMAD ET AL.